Model production for animation libraries

• US 8,144,153 B1
• Filed: 11/20/2007
• Issued: 03/27/2012
• Est. Priority Date: 11/20/2007
• Status: Active Grant

• Current Assignee: Lucasfilm Limited (The Walt Disney Company)
• Original Assignee: Lucasfilm Limited (The Walt Disney Company)
• Inventors: Sullivan, Steve, Callari, Francesco
• Primary Examiner(s): Broome, Said
• Application Number:
  US11/943,556
• Time in Patent Office: 1,589 Days
• Field of Search: None
• US Class Current: 345/473
• CPC Class Codes: G06T 13/40 of characters, e.g. humans, animals or virtual beings
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 2207/30204 Marker
• CPC Class Codes: G06T 7/251 involving models

Abstract

A computer-implemented method includes selecting a subset of images from a set of captured images. A surface feature of one object is represented in the content of each selected subset image. The method also includes decomposing the surface feature content of each selected image to produce a model of representations of the object.

Claims

1. A computer-implemented method comprising:

selecting, in a processor, a subset of images from a set of captured images, wherein a surface feature of a first object is represented in content of the selected image subset;

producing a model, at a first time, of at least one representation of the first object by decomposing the content of the selected image subset into principal components;

using the model to track surface features in at least one image that is separate from the selected image subset used to produce the model, wherein tracking the surface features provides an estimate of the surface features represented in the at least one image that is separate from the selected image subset used to produce the model; and

combining with the model, at a second time, content that represents the estimated surface features from the tracked surface features of the at least one separate image by decomposing content into principal components to produce an enhanced model of additional representations.

2. The computer-implemented method of claim 1, further comprising:

storing the enhanced model in a library of models.

3. The computer-implemented method of claim 1, in which combining the content includes decomposing the content with a linear transformation.

4. The computer-implemented method of claim 3, in which decomposing the content with a linear transformation includes computing principle components.

5. The computer-implemented method of claim 1, in which the image subset selection is based upon the surface feature represented in the content of the subset images.

6. The computer-implemented method of claim 1, in which the at least one separate image includes content that represents a second object, different from the first object.

7. The computer-implemented method of claim 1, in which the at least one separate image includes content that represents the first object.

8. The computer-implemented method of claim 1, in which the content is represented in a mesh.

9. The computer-implemented method of claim 1, in which the surface feature includes an artificial surface feature.

10. The computer-implemented method of claim 1, in which the surface feature includes a natural surface feature.

11. A system comprising:

a computer system comprising;

a memory for storing instructions of a model initiator and a model updater;

a processor for executing instructions;

the model initiator to select a subset of images from a set of captured images, wherein a surface feature of a first object is represented in content of the selected image subset, the model initiator also decomposing the content of the selected image subset into principal components to produce a model, at a first time, of at least one representation of the first object; and

the model updater to use the model to track surface features in at least one image that is separate from the selected image subset used to produce the model, wherein tracking the surface features provides an estimate of the surface features represented in at least one image that is separate from the selected image subset used to produce the model, the model updater is further configured to combine, at a second time, the model and content that represents the estimated surface features from the tracked surface features of the at least one separate image by decomposing content into principal components to produce an enhanced model of additional representations.

12. The system of claim 11, in which the model initiator is capable of storing the model in a library of models and the model updater is capable of storing the enhanced model in the library of models.

13. The system of claim 11, in which the model updater is configured to compute a linear decomposition to combine the content with the model.

14. The system of claim 13, in which the linear decomposition computation includes computing principle components.

15. The system of claim 11, in which the image subset selection is based upon the surface feature represented in the content of the subset images.

16. The system of claim 11, in which the at least one separate image includes content that represents a second object, different from the first object.

17. The system of claim 11, in which the at least one separate image includes content that represents the first object.

18. A computer program product tangibly embodied in a storage device and comprising instructions that when executed by a processor perform a method comprising:

selecting a subset of images from a set of captured images, wherein a surface feature of a first object is represented in content of the selected image subset;

producing a model, at a first time, of at least one representation of the first object by decomposing content of the selected image subset into principal components;

using the model to track surface features in at least one image that is separate from the selected image subset used to produce the model, wherein tracking the surface features provides an estimate of the surface features represented in at least one image that is separate from the selected image subset used to produce the model; and

combining, at a second time, the model and content that represents the estimated surface features from the tracked surface features of the at least one separate image by decomposing content into principal components to produce an enhanced model of additional representations.

19. The computer program product of claim 18, further comprising instructions that when executed by a processor perform a method comprising:

storing the enhanced model in a library of models.

20. The computer program product of claim 18, in which combining the content includes decomposing the content with a linear transformation.

21. The computer program product of claim 20, in which decomposing the content with a linear transformation includes computing principle components.

22. The computer program product of claim 18, in which the image subset selection is based upon the surface feature represented in the content of the subset images.

23. The computer program product of claim 18, in which the at least one separate image includes content that represents a second object, different from the first object.

24. The computer program product of claim 18, in which the at least one separate image includes content that represents the first object.

25. The computer program product of claim 18, in which wherein the content is represented in a mesh.